A pioneering cryptic random projection based approach for privacy preserving data mining

Authors:
K. Murugesan;Md. Rukunuddin Ghalib;J. Gitanjali;J. Indumathi;D. Manjula
Affiliations:
Department of Computer Science and Engineering, Anna University, Chennai;Department of Information Technology, Vellore Institute of technology, Vellore,Tamilnadu, India;Department of Information Technology, Vellore Institute of technology, Vellore,Tamilnadu, India;Department of Computer Science and Engineering, Anna University, Chennai;Department of Computer Science and Engineering, Anna University, Chennai
Venue:
IRI'09 Proceedings of the 10th IEEE international conference on Information Reuse & Integration
Year:
2009

Citing 5
Cited 0

Database-friendly random projections

PODS '01 Proceedings of the twentieth ACM SIGMOD-SIGACT-SIGART symposium on Principles of database systems
Privacy-preserving Distributed Clustering using Generative Models

ICDM '03 Proceedings of the Third IEEE International Conference on Data Mining
Very sparse random projections

Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining
A novel framework for optimised privacy preserving data mining using the innovative desultory technique

International Journal of Computer Applications in Technology
Improving random projections using marginal information

COLT'06 Proceedings of the 19th annual conference on Learning Theory

Quantified Score

Hi-index	0.00

Visualization

Abstract

Privacy is the most important apprehension in many data mining applications. In this paper a new technique called Cryptic Random Projection, solves the re-identification quandary (which is found in the conventional random projections). Here this encryption based random projection assigns secret keys to the positions of random matrix elements and not to the random numbers. We have addressed two kinds of random sequences for generating the random sequences called determinist and indeterminist random sequences and encrypted it in a new way so that the original data cannot be re-identified. We have also optimized the privacy level which toughens the re-identification of original data without compromising the processing speed and data utility. We hope the projected solution will tarmac way for investigation track and toil well according to the evaluation metrics including hiding effects, data utility, and time performance.